Conventional data recovery technologies emphasize backup and storage. These include making a copy of data and storing the copy version at a different location from the original version. This creates an inefficient use of data since two copies of data are retained.
There have been several advancements in backup and storage technology. One such advancement is deduplication. Deduplication removes redundant data blocks so that only one instance of the data block is stored. This can save potentially massive amounts of storage space, as well as bandwidth when transferring data from a backup client to a backup server.
Though such advancements in backup and storage have led to some increased efficiency, much of the efficiency is lost when recovering the data. For example, when recovering from deduplicated data, the data is reconstituted first and then sent to the recovery target. The recovery process requires much more bandwidth resources than the backup process.
There is a need, therefore, for an improved method, article of manufacture, and apparatus for recovering data in data systems.